An information theory approach to quantifying the sequence-dependent response of nucleic acid motors with applications to nanopore DNA sequencing

Motor enzymes that interact with DNA are essential for replicative biological processes. In nanopore sequencing, a motor enzyme controls the motion of a nucleic acid through a protein nanopore, and sequence-dependent blockages of an ion current flowing through the nanopore are used to decode the DNA sequence. The kinetics of these enzymes are sequence-dependent and can serve as an additional source of information during sequencing. Here, we use Mutual Information (MI) to quantify the sequence-dependent kinetics of a Hel308 helicase during nanopore sequencing. We use MI to identify sites in Hel308 that are responsible for sequence-dependent kinetics and develop "k-mer" models of Hel308 kinetics that map kinetics to DNA sequence. We estimate that enzyme kinetics can improve nanopore sequencing accuracy by ~5-fold at high sequencing depth. We mutate Hel308 to identify amino acids involved in DNA translocation and suggest pathways for engineering molecular motors with enhanced responsiveness to DNA sequence.